A Bayesian approach for predicting functional reliability of one-shot devices

Byeong Min Mun, Chinuk Lee, Seung gyo Jang, Byung Tae Ryu, Suk Joo Bae

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Accelerated life tests (ALTs) have been used to assess reliability of one-shot devices in a short time. Due to destructive characteristics of one-shot devices, lifetime data of the devices is incomplete and enough number of failures or even no failures may be not secured in ALT. In such situations, Baysian methods incorporating prior information into the parameters provide useful inference on the reliability of one-shot devices. In this paper, we propose a modeling approach to predict functional reliability of pin pullers as a kind of one-shot devices, mainly in a Bayesian framework. We introduce three different priors to the parameters of the Weibull distribution or reliability function. Sress-strength relationships of key components in pin pullers are employed to the scale and shape parameters via three prior densities. The proposed methods are illustrated with a variety of simulation studies. The simulation works are performed using the Gibbs sampling technique to generate MCMC samples to obtain Bayesian estimates of the Weibull parameters. The Bayesian estimates from the three priors tend to approach to true parameter values as sample size increases.

Original languageEnglish
Pages (from-to)71-82
Number of pages12
JournalInternational Journal of Industrial Engineering : Theory Applications and Practice
Volume26
Issue number1
StatePublished - 2019 Jan 1

Fingerprint

Weibull distribution
Sampling

Keywords

  • Bayesian approach
  • Functional reliability
  • One-shot device
  • Pin puller
  • Weibull distribution

Cite this

@article{4419ea9c86ce4e548948d0cb1d9e3fa0,
title = "A Bayesian approach for predicting functional reliability of one-shot devices",
abstract = "Accelerated life tests (ALTs) have been used to assess reliability of one-shot devices in a short time. Due to destructive characteristics of one-shot devices, lifetime data of the devices is incomplete and enough number of failures or even no failures may be not secured in ALT. In such situations, Baysian methods incorporating prior information into the parameters provide useful inference on the reliability of one-shot devices. In this paper, we propose a modeling approach to predict functional reliability of pin pullers as a kind of one-shot devices, mainly in a Bayesian framework. We introduce three different priors to the parameters of the Weibull distribution or reliability function. Sress-strength relationships of key components in pin pullers are employed to the scale and shape parameters via three prior densities. The proposed methods are illustrated with a variety of simulation studies. The simulation works are performed using the Gibbs sampling technique to generate MCMC samples to obtain Bayesian estimates of the Weibull parameters. The Bayesian estimates from the three priors tend to approach to true parameter values as sample size increases.",
keywords = "Bayesian approach, Functional reliability, One-shot device, Pin puller, Weibull distribution",
author = "Mun, {Byeong Min} and Chinuk Lee and Jang, {Seung gyo} and Ryu, {Byung Tae} and Bae, {Suk Joo}",
year = "2019",
month = "1",
day = "1",
language = "English",
volume = "26",
pages = "71--82",
journal = "International Journal of Industrial Engineering : Theory Applications and Practice",
issn = "1072-4761",
number = "1",

}

A Bayesian approach for predicting functional reliability of one-shot devices. / Mun, Byeong Min; Lee, Chinuk; Jang, Seung gyo; Ryu, Byung Tae; Bae, Suk Joo.

In: International Journal of Industrial Engineering : Theory Applications and Practice, Vol. 26, No. 1, 01.01.2019, p. 71-82.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - A Bayesian approach for predicting functional reliability of one-shot devices

AU - Mun, Byeong Min

AU - Lee, Chinuk

AU - Jang, Seung gyo

AU - Ryu, Byung Tae

AU - Bae, Suk Joo

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Accelerated life tests (ALTs) have been used to assess reliability of one-shot devices in a short time. Due to destructive characteristics of one-shot devices, lifetime data of the devices is incomplete and enough number of failures or even no failures may be not secured in ALT. In such situations, Baysian methods incorporating prior information into the parameters provide useful inference on the reliability of one-shot devices. In this paper, we propose a modeling approach to predict functional reliability of pin pullers as a kind of one-shot devices, mainly in a Bayesian framework. We introduce three different priors to the parameters of the Weibull distribution or reliability function. Sress-strength relationships of key components in pin pullers are employed to the scale and shape parameters via three prior densities. The proposed methods are illustrated with a variety of simulation studies. The simulation works are performed using the Gibbs sampling technique to generate MCMC samples to obtain Bayesian estimates of the Weibull parameters. The Bayesian estimates from the three priors tend to approach to true parameter values as sample size increases.

AB - Accelerated life tests (ALTs) have been used to assess reliability of one-shot devices in a short time. Due to destructive characteristics of one-shot devices, lifetime data of the devices is incomplete and enough number of failures or even no failures may be not secured in ALT. In such situations, Baysian methods incorporating prior information into the parameters provide useful inference on the reliability of one-shot devices. In this paper, we propose a modeling approach to predict functional reliability of pin pullers as a kind of one-shot devices, mainly in a Bayesian framework. We introduce three different priors to the parameters of the Weibull distribution or reliability function. Sress-strength relationships of key components in pin pullers are employed to the scale and shape parameters via three prior densities. The proposed methods are illustrated with a variety of simulation studies. The simulation works are performed using the Gibbs sampling technique to generate MCMC samples to obtain Bayesian estimates of the Weibull parameters. The Bayesian estimates from the three priors tend to approach to true parameter values as sample size increases.

KW - Bayesian approach

KW - Functional reliability

KW - One-shot device

KW - Pin puller

KW - Weibull distribution

UR - http://www.scopus.com/inward/record.url?scp=85067951089&partnerID=8YFLogxK

M3 - Article

VL - 26

SP - 71

EP - 82

JO - International Journal of Industrial Engineering : Theory Applications and Practice

JF - International Journal of Industrial Engineering : Theory Applications and Practice

SN - 1072-4761

IS - 1

ER -